WO1985003948A1

WO1985003948A1 - Macrophage activating factor

Info

Publication number: WO1985003948A1
Application number: PCT/GB1985/000088
Authority: WO
Inventors: Ann Dorothy Mary Rees
Original assignee: Celltech Limited
Priority date: 1984-03-02
Filing date: 1985-03-04
Publication date: 1985-09-12
Also published as: JPS61501563A; GB8405555D0; EP0175705A1

Abstract

A macrophage activating factor which is antigenically distinct from interferon - gamma. Such macrophage activating factors are produced by culturing cloned human T cell lines TB 68.2.1, TB 68.2.2 and TB 68.2.6.

Description

Macrophage Activating Factor Field of the Invention

This invention relates to a macrophage activating factor (referred to herein also as MAF) and to cell lines producing same. Background to the Invention

Antigen or mitogen stimulated T lymphocytes screte a heterogeneous mixture of substances , known collectively as macrophage activating factors which induce macrophages to kill tumours, bacteria, fungi and parasites. T lymphocytes produce interferon-γ (IFN-γ ) which is known to activate macrophages and is the only identified and characterised macrophage activating factor.

We have now discovered a macrophage activating factor which is antigenically distinct from interferon-γ . In particular we have cloned three human T cells that are specific for a monoclonal antibody-derived antigen for Mycobacterium tuberculosis (M tuberculosis) and have shown that supernatant from these cell lines contains at least one macrophage activating factor which is antigenically distinct from interferon-γ and can be separated from interferon-γ by high performance liquid chromatography. Summary of the Invention

According to the present invention we provide a macrophage activating factor characterised in that the macrophage activating factor is antigenically distinct from interferon-γ. Preferably the macrophage activating factor is separable from interferon-γ by high performance liquid chromatography. Preferably the macrophage activating factor is antigenically the same as or similar to a macrophage activating factor produced by cloned human T cell line TB 68.2.1 (referred to later as Clone 1), TB 68.2.2 (referred to later as Clone 2) or TB 68.2.6 (referred to later as Clone 6) .

The cloned human T cell lines were deposited on 4 March 1985 at the National Collection of Animal

Cell Cultures, PHLS Centre for Applied Microbiology and Research, Porton Down, Salisbury, Wiltshire SP4 OJG, United Kingdom under accession numbers 85030401 (TB 68.2.1), 85030402 (TB 68.2.2) and 85030403 (TB 68.2.6). These deposited cell lines are cloned human T cells produced by antigen stimulation of peripheral blood mononuclear cells. As such their physical characteristics closely resemble similar cloned and cultured human T cells , the cell lines being characterised by the excretion into their culture supernatant of a macrophage activating factor which is antigenically distinct from interferon-γ .

In another aspect of the invention we provide a cloned T cell line selected from TB 68.2.1, TB 68.2.2 and TB 68.2.6 and mutants and derivatives thereof. As used herein the term 'mutants and derivatives thereof includes any human T cell line derived from a cell line as defined above which exhibits the characteristic of producing a macrophage activating factor which is antigenically distinct from interferon- γ.

Macrophage activator factor according to the invention may be prepared by cloning T cells which have been prestimulated with an appropriate antigen. We have discovered that a purified antigen of

M. tuberculosis (hereinafter referred to as the TB 68 antigen) prepared from immunoadsorbants coupled to a monoclonal antibody TB 68 is particularly useful for generating T cell clones. Other antigenic preparations which contain the TB 68 antigen (for example M. tuberculosis strain H37v, M.bovis strain BCG and tuberculin purified protein derivative (PPD)) may also be used. We further provide a process for producing macrophage activating factor of the invention comprising the steps of culturing a cloned human T cell which has been stimulated prior to cloning with a TB 68 antigen and isolating the macrophage activating factor from the culture supernatant. Preferably the cell lines are those designated TB 68.2.1, TB 68.2.2 and TB 68.2.6 referred to above.

Preferably the cloned T cell line is cultured in a culture medium which includes a cell growth factor such as, for example, interleukin 2.

Preferably the cloned human T cell line is cultured in the presence of an antigen which stimulates secretion of macrophage activating factor, such as, for example, the TB 68 antigen and autologous feeder cells such as, for example, fresh identical HLA DR-identcal or part identical cells (monocytes/macrophages). The feeder cells act to present the antigen to the growing cell line.

Alternatively the cloned human T cell line is cultured in the presence of a monoclonal antibody having specificity for the T3 cell surface antigen. A suitable such monoclonal antibody is that designated OKT-3 (Ortho Diagnostics Systems Inc.).

The macrophage activating factor of the invention has therapeutic uses in the treatment of tumours and the prophylaxis or treatment of infection, for example, by bacteria, fungi or parasites. We further provide macrophage activating factor of the invention for use as a pharmaceutical, preferably in the applications previously described. We also provide a pharmaceutical composition comprising macrophage activating factor of the invention and a pharmaceutically acceptable excipient. We also provide a process for preparing such a composition comprising bringing macrophage activating factor of the invention into association with a pharmaceutically acceptable excipient. We further provide a method of treatment of tumours and a method of prophylaxis or treatment of infection, for example by bacteria, fungi or parasites comprising administering an effective amount of a macrophage activating factor of the invention. Brief Description of the Drawings

Figure 1 shows a graph indicating the enhancement of hydrogen peroxide release from human macrophage-like cells by recombinant DNA-derived interferon-γ (circles) , native interferon-γ (squares) and supernatant from a T cell clone (triangles).

Figure 2 shows the fractionation of macrophage activating factor from different sources. Detailed Description of Embodiments

EXAMPLE 1

PREPARATION OF THE TB 68 ANTIGEN

1) Preparation of Immunoadsorbant

2 mg/ml monoclonal antibody TB 68 (Coates et al.. , 1982) was coupled to protein A sepharose beads (Sigma, London) by rotation overnight at 4°C. The beads were then washed first with phosphate buffered saline 3 times and then with 0.1 M sodium borate buffer saline pH 9.0 (x3) . The beads were resuspended at a concentration of 10 ml of this buffer to 1 ml beads. 20 mM dimethyl-pimelimidate dihydrochloride was then added and incubated for 1 hour at room temperature. The beads were then washed (x3) in the borate buffer and incubated for 5 minutes at RT in 0.02 M ethanolamine. The beads were washed with PBS and packed into 2 ml columns. The level of coupling was checeked by mock elution with 8 M urea.

2) Running the Column

2 ml of M tuberculosis strains H37Rv express (Coates et al . , 1981) was run slowly onto a 2 ml column (1 ml/h) . This was then slowly washed, at the same rate, with 2 ml of PBS. The column was then washed more rapidly with PBS. The column was then washed more rapidly with 8 M urea (1 ml/h^-1) 1 ml fractions were collected. Fractions 2-5 were combined and dialysed against 5 L of PBS followed by 10 L of dH₂O using a continous dialysis system. The eluate, containing the antigen was then concentrated by immersion in Sephadex G200 to a final volume of approximately 1 ml. Protein concentration was then determined by Lowry's method. 3) Recovery and Characteristics of Antigen

Yield of antigen was estimated to be 50 μg/10 mg of expressate (approx 0.5% of total protein in M tuberculosis strain H37Rv expressate).

Immunoblotting indicates the native molecular weight of the antigen to be 120 K daltons.

8 M urea treatment reduces this protein to 4 bands of molecular weights of 69, 46, 43 and 30 K daltons. These bands are immunoprecipitated by monoclonal antibody TB 68 and not by any other monoclonal antibody.

EXAMPLE 2

CLONING OF THE T CELL LINES TB 68.2.1, TB 68.2.2

AND TB 68.2.6

Methods and Materials

Preparation of Antigen Activated PBMC

Peripheral blood mononuclear cells (PBMC) were obtained by ficollhypague centrifugation as described by Boyum, 1968. PBMC were adjusted to a concentration of 2 x 10⁶ cells/ml in maintenance medium RPMI 1640 (Gibco biocult Limited) with 20% autologous serum (heat activated) and incubated with antigen for 4 days at 37°C in 5% CO₂ in air. 200 ml of cells were incubated with 20 ml of 0.0117 g/ml TB 68 antigen. After 4 days in culture the volume of maintenance medium was doubled and 15% IL2 added. The cells were then incubated for a further 2 days before harvesting for cloning.

Cloning of Antigen Activated Lymphocytes

After a total of 6 days in culture in the presence of antigen, cells were harvested, resuspended over ficoll hypague in 12 x 75 mm sterile capped tubes (Falcon Plastic Limited) and centrifuged for 20 minutes at 200 x G. Cells at the interphase were enriched 2 fold for lymphoblasts and comprised 50% of the cells counted. Blast enriched suspensions were diluted to 33^1/3 cells/ml of maintenance medium containing 15% of IL2 and 20% autologous serum and 10 μl was plated in sterile 60 well microterasaki plates (Nunc Limited). 10⁴ autologous cells per well were gamma irradiated (3,000 rad, ¹³⁷Cs) and added with antigen at optimal concentration to the wells containing blasts in 10 μl aliquots. The optimal concentration of serum and IL2 was estimated from proliferation assay. Cultures were incubated for 7 days in humidified chambers at 37°C in 5% CO₂/air. After this plates were scanned microscopically for wells containing clones. Growing cells were then transferred to 96 well flat-bottomed microtitre plates (Flow Laboratories) containing fresh maintenance medium supplemented with 5 x 10⁵/ml autologous irradiated PBMC (feeders), 15% IL2 and 0.0117 g/ml TB 68 antigen. After 7 additional days in culture the clones were transferred to 24 well trays (Linbro Scientific companies) containing the appropriate amounts of IL2, autologous feeders and TB 68 antigen in a total volume of 2 ml. Cultures received fresh IL2 every 3 to 4 days, alternating with irradiated autologous feeders 5 x 10⁵/ml and antigen and were thus maintained throughout the course of the experiments. Clones were allowed to grow for 4 days after the addition of IL2 and feeders before testing in proliferative assays. The above procedure gave rise to 3 cloned T cell lines designated TB 68.2.1, TB 68.2.2 and TB 68.2.6.

Antigens

TB 68 antigen was prepared as described in Example 1. TB 72 antigen was similarly prepared.

Other antigenic preparations of mycobacteria used were mycobacterial pressates; the non-particulate fraction of a bacterial homogenate passed through a press as described by Hewitt e t al.. , 1982. Tuberculin purified protein derivative (PPD) was obtained from Evans Medical Supplies Limited) .

Interleukin 2

This was obtained from Biotest Folex Limited as 'Lymphocult T, lectin free containing less than 0.1% phytohaemagglutinin by filtration. Optimal concentrations of IL2 for the maintenance of clones was determined by titration against the clone donor PHA blasts. Proliferative Assays

Proliferative responses were determined using a microterasaki method as described by O'Brien and Knight et. al. , 1979. Briefly, cloned cells at 2.5 x 10⁵/ml in maintenance medium with 20% heat inactivated autologous serum were added in 20 μl aliquots to microterasaki plates containing antigen in 2 1.

In the following Examples, 3 and 4, the clones referred to as Clone 1, 2 and 6 are the Clones TB 68.2.1. TB 68.2.2 and TB 68.2.6 respectively. The clones referred to as Clones 7 and 13 in Example 4 are not included within the scope of the invention.

EXAMPLE 3

DEMONSTRATION THAT CLONES TB 68.2.1, TB 68.2.2 AND TB 68.2.6 SECRETE MAF WHICH IS ANTIGENICALLY DISTINCT FROM IFN-γ

Antigen or mitogen stimulated T lymphocytes secrete macrophage activating factors (MAF) which induce macrophages to kill tumours ^1,2 bacteria^3,4, fungi⁵ and parasites^6,7. These lymphocytes also produce interferon-γ (IFN-γ) ⁸. A crucial question is whether macrophage activation by T cell supernatants is due entirely or partly to IFN-γ. Both human and murine IFN-γ have been produced in high purity by recombinant DNA techniques ^{1 1-1 3}. However, the existence of MAF antigenically distinct from IFN-γ in T cell supernatants has not been established. We have used an anti-IFN-γ monoclonal antibody (MAB) to test for MAF other than IFN-γ in the culture supernatants of three human T cell clones (prepared as described in Example 2) that are specific for a MAB-derived antigen from Mycobacterium tuberculosis. We have found MAF that is antigenically distinct from IFN-γ and can be separated from IFN-γ by high performance liquid chromatography (HPLC) .

Although IFN-γ activates macrophages to kill intracellular parasites, to lyse tumour cells and to release the microbicidal agent hydrogen peroxidase⁶ it is not known whether srtigen-stimulated T lymphocytes secrete MAF distinguishable from IFN-γ. One hypothesis is that IFN-γ is one of several MAFs which may function independently or synergistically. The finding of MAF activity but not IFN-γ activity in supernatants from murine T cell clones² and murine T cell hybridomas^14,15 suggests the existence of several MAFs.

With the expression of human IFN-γ in Escherichia coli⁹. pure IFN-γ has become available with which macrophages can be activated without contamination with other lymphokines which occur in T cell supernatants. A monoclonal antibody (MAB 12.20; Genentech Inc) has been produced which neutralises the cloned IFN-γ and natural IFN-γ. This provided a means to search human T cell supernatants for MAF activity that is unaffected by this MAB.

Cells of the human macrophage-like line U937 can be immunologically activated to express anti-microbial activity¹⁶ and we used enhancement of hydrogen peroxide releasel? from these cells to assay MAF. We compared the activation induced by cloned IFN-γ and supernatants from three T cell clones as prepared in Example 2 that proliferated in the presence of a M tuberculosis antigen purified by affinity chromatography (as described in Example 1) with the MAB TB 68 described by Coates et al. ¹⁸ (Figure 1). Both T cell supernatants and cloned IFN-γ activated the cells for peroxide release but the T cell supernatants were about 50-fold more potent than would be expected if their activity was solely due to their interferon content. Anti-IFN-γ MAB 12:20 obliterated the effect of up to 100 units of cloned IFN-γ. Natural IFN-γ also activated U937 cells for peroxide release and this activity too was substantially reduced by MAB 12:20. In contrast to IFN-γ, the MAF activity of the T cell clone supernatants was only slightly reduced by MAB 12:20, a finding compatible with the neutralisation of the 10-60 units interferon (anti-viral) activity per ml in these supernatants.

These results indicate a MAF in T cell clone supernatants that is antigenically distinct from IFN-γ. The neutralisation of the activating effect of IFN-γ by MAB 12:20 was by specific binding of this antibody to IFN-γ and was not caused by interaction of antibody with F_c receptors on the U937 cells as MAB NB10 (Celltech Limited), an antibody with the same isotype as 12:20 but unrelated specificity, had no effect on peroxide release (Figure 1) . The possibility that inhibitors of IFN-γ or inhibitors of binding of anti-IFN-γ to IFN-γ were present in T cell clone supernatants is unlikely since, in preliminary experiments, cloned IFN-γ added to the supernatants was fully effective in the MAF assay and was neutralised as expected by another anti-IFN-γ (MAB 12:09; Genentech Inc). The activity we describe was not dependent on using the myeloma (U937) as detector cell; the T cell clone supernatants also enhanced peroxide release from conventional macrophages matured from human blood. Physiochemical characterisation of a supernatant from a murine T cell line EL-4 has revealed two MAFs, one of which is distinct from IFN-γ and is unaffected by anti-IFN-γ serum^19,20. In our experiments, molecular filtration by HPLC revealed that the MAF in T cell clone supernatants was distinguishable from IFN-γ (Figure 2) . Activity in supernatant from T cell Clone 1, for example, appeared in two regions, at 24,000 and 36,000 daltons. This contrasts with activity in supernatants from phytohaemagglutinin-stimulated human spleen cells which was at 51,000 daltons and cloned IFN-γ at 78,000 daltons. This indicates a MAF which is not only antigenically distinct from cloned IFN-γ, but also has a different molecular weight. Since the activity in a human spleen cell supernatant had a molecular weight distinct from either IFN-γ or T cell clone supernatants, the likelihood of multiple molecular forms of MAF is increased. Reported molecular weights for natural IFN-γ range from 35,000-70,000^21-24. IFN-γ consists of subunits of 17,126 daltons in the case of recombinant IFN-γ⁹ or of 20,000 and 25,000 daltons²⁵. The MAF described here seems unlikely to be a monomer and dimer of IFN-γ subunits because it is antigenically distinct from both cloned and natural IFN-γ and MAB 12:20 binds to both monomer and dimer of cloned IFN-γ (Genentech Inc, unpublished). Further characterisation of the seemingly novel MAF(s) produced by these human T cell clones is in progress. LEGENDS

Figure 1

Enhancement of hydrogen peroxide from human macrophage-like cells by exposure to T cell clone supernatant or IFN-γ and the effects of monoclonal anti-IFN-γ antibody. T cell clone supernatant was collected during clone maintenance on irradiated feeder cells with interleukin 2 and antigen²⁶. From an exponentially replicating culture of human macrophage-like lymphoma cell line U937¹⁶, in RPMI 1640 medium containing 10% foetal calf serum, aliquots of 5 x 10⁵ cells were dispensed into plastic tubes (12 x 75 mm; Falcon). The tubes were gently centrifuged and the supernatants removed. Dilutions of T cell clone supernatant or IFN-γ were made in U937 growth medium and added to the tubes either directy or after mixing with antibody (100 μg ml^-1) and incubating for 2 hours at 37°C. Each tube contained a final volume of 500 μl. The tubes were incubated for 3 days at 37°C with daily centrifugation and replacement of medium and supplements. Hydrogen peroxide release from the cells during 1-h stimulation with phorbol myristate acetate (1 μg ml^-1) was then measured and the number of cells in each tube assessed by DNA estimation¹⁷. Controls included tubes containing either no supplements or sham supernatants prepared identically to T cell clone supernatants but without using T cells. Peroxide release in these controls (2-6 nmol 10⁰ cells^-1) was subtracted from the relevant test values and the difference was plotted as increment in peroxide release versus units of interferon in the test supplements. Interferon was assayed as inhibition of nucleic acid synthesis in SFV-infected WISH fibroblast cells²⁷. Illustrated are results obtained in a representative experiment which included recombinant DNA-derived IFN-γ (o,●), native IFN-γ (from PHA-stimulated human blood lymphocytes, Finnish Red Cross;♢ ,♦ ) , supernatant from T cell Clone 1 (Δ, ▲) used either alone (open symbols) or after treatment with anti-IFN-γ MAB 12:20 C₈ (closed symbols). Resnlts with recombinant DNA-derived IFN-γ treated with MAB NB10 are also shown Supernatants from 3 discrete T cell clones (1, 2 and

6) gave similar results and for clarity only data from one of these is shown. Regressions were fitted by the least squares method.

Figure 2

Fractination of macrophage activity from different sources. Supernatants were concentrated 5 to 10-fold by lyophilisation then up to 2 ml was applied to a gel filtration column (21.5 x 600 mm, Ultropac TSE. 3,000 SW; LKB Limited) and eluted with 0.1 M sodium phosphate buffer, pH 7.0, containing 0.1 M NaCl at a flow rate of 5.1 ml min^-1. MAF activity in the fractions was measured by increment in peroxide release from U937 cells as described in Figure 1. The chromatograms shown are from three separate experiments using supernatants from PHS-stimulated human spleen cells (o), from T cell Clone 1 (●) and cloned IFN (A). Supernatants from T cells Clones 2 and 6 yielded chromatograms resembling that from Clone 1 and the data are omitted for clarity. Molecular weight marker positions are indicated at the top of the Figure. References

1) Kelso, A; Glasebrook, A L; Kanagawa, 0 and Brunner, K T. J. Immunol

129. 550-556 (1982)

2) Gemsa. D; Debatin, K M; Kramer, W; Kubelka. C; Deimann, W; Kees, U and Krammer, PH. J. Immunol (in press)

3) Simon, H B and Sheagren, J N. Cell Immuno 4., 163-174 (1972)

4) Walker, L and Lowrie. D B. Nature 293. 69-70 (1981)

5) Sasada, M and Johnson. R B. J Exn. Med 152. 85-98 (1980)

6) Nathan, C F; Noguiera, N; Juangbhanich, C W; Ellis, J and Cohn, Z J

Exp Med 149, 1056-1068 (1979)

7) Buchmuller, Y and Mauel. J. J Immunol 10, 428-433 (1980)

8) Marcucci, F; Waller, M; Kirchner, H and Krammer, P. Nature 291. 79-81 (1981)

9) Gray, P W; Leung, D W; Pennica, D; Yelverton, E; Najarian. R; Simonsen, C C; Derynck, R; Sherwood, P J; Wallace, D M; Berger, S L; Levinson. A D and Goeddel. D V. Nature 295. 503-509 (1982)

10) Gray, P W and Goeddel, D V Proc Nat Acad Sci USA (in press)

11) Nathan, C F; Murray, H W; Wiebe, M E and Rubin. B Y J Exp Med

158. 670-689 (1983)

12) Pace. J L; Russell, S W; Torres, B A, Johnson, H M and Gray, P W. J

Immunol 130. 2011-2013 (1983)

13) Schultz, R M and Kleinschmidt, W J. Nature 305. 239-240 (1983)

14) Ratliff, T L; Thomasson, D L; McCool, R E and Catalona. W J. Cell

Immunol 68, 311-321 (1982)

15) Schreiber, R D; Altman, A and Katz, D H. J Exp Med 156. 677- 689 (1982)

16) Wing, E J; Koren, H S; Fischer. D G and Kelley, V. J Reticuloendoth Soc 29, 321-328 (1981)

17) Jackett. P S; Andrew. P W; Aber, V R and Lowrie. D B. B J Exp

Path 62, 419-428 (1981)

18) Coates, A R M; Hewitt, J; Allen, B W; Ivanyi, J and Mitchison. D A.

Lancet ii, 167-169 (1981)

19) Meltzer. M S; Benjamin, W R and Farrar, J J. J Immunol 129. 2802-2807 (1982) 20) Nacy, C A; James, S L; Benjamin, W R; Farrar, J J; Hockmeyer, W T and Meltzer, M S. Infect Immun 40, 820-824 (1983)

21) Langford, M P; Gregiades, J A; Stanton, G J; Dianzani, F and Johnson, H M. Infect Immun 26, 36-41 (1979)

22) DeLey, M; van Damme, J; Claeys, H; Weening, H; Heine, J W; Billiaa, A; Vermylen, C and de Somer, P. Eur J Immunol 10. 877-883 (1980)

23) Yip, Y K; Pang, R H L; Oppenheim, J D; Nachbar. M S; Henriksen, D; Zerebeckyj-Eckhardt. I and Vilcek. J. Infect Immun 34, 131-139 (1981)

24) Pestka, S; Eelder, B; Familletti, P C; Moschera, J A; Crowl, R and Kempner. E S. J Biol Chem 258. 9706-9709 (1983)

25) Yip, Y K; Barrowclough, B S; Urban. C and Vilcek, J. Science 215. 411-413 (1982)

26) Bach, F H; Bunzendahl. H; Wee, S L; Iwaki, J and Segall. M in

Isolation. Characterisation, and utilization of T Lvmophocvte Clones (eds Fathman. C G and Fitch, F W) 413-424 (Academic, New York, 1982)

27) Atherton. K T and Burke. D C. J Gen Virol 20, 297-304 (1975)

EXAMPLE 4

CHARACTERISATION OF CLONED T CELL LINES TB 68.2.1, TB 68.2.2 AND TB 68.2.6

It is clear from studies of host responses to facultative intracellular parasites which has been particularly studied in the mouse^1,2 that anti-microbial immunity involves the collaboration of specific lymphocytes, or their products, to activate macrophages. Whilst lymphocytes respond with immunological specificity to the invading microorganism little is known of the importance of individual antigens. It is not known, for example, whether one antigen may preferentially generate one particular, or several of the responses known to follow stimulation wtih bacterial or their antigens. We have, therefore, used a purified antigen³ of Mycobacterium tuberculosis prepared from immunoadsorbants coupled to a monoclonal antibody (MAB)⁴ TB 68 to generate T cell clones (as described in Examples 1 and 2). Using these clones we have examined part of the response to this organism. The majority of clones wer helper/inducer T cells but one clone is described which had a suppressor/cytotoxic phenotype. Clones were generated that were specific for the eliciting antigen. Helper/inducer cloned T cells constituitively secreted lymphokines that were able to activate macrophages for the enhanced release of H₂O₂, an oxygen derivative implicated in macrophage anti-microbial activity⁵.

The clones described were prepared from a single prestimulation of normal peripheral blood mononuclear cells (PBMC) with TB 68 antigen, by a modification⁶ of a method previously described⁷. Plating efficiency, (the proportion of cells proliferating) was estimated to be 20% and, therefore clonality was established by subcloning. Plating efficiency at subcloning was 95% and subclones were, by our criteria, identical to the parent clone. The clones generated were screened initially for the capacity to respond to the eliciting antigen. The majority of the clones (75%) failed to respond. Of the 25% which did respond 5 of the clones were selected for futher investigation. The phenotype of these clones is described in Table 1. It can be seen that 4 of the clones were positive with the monoclonal antibodies 3a and 8a (Beckton Dickinson, California, USA), a phenotype consistent with helper/inducer T lymρhocytes⁸. Clone 7, uniquely, was positive with monoclonal antibody 2a and 8a consistent with a suppressor/cytotoxic phenotype⁸. All of the clones were positive with leu 1, which is present on 95% of peripheral blood T lymphocytes⁸.

The antigenic specificity of individual clones was determined by co-culturing clones with irradiated autologous PBMC and a variety of antigenic preparations some of which were known to contain, at lease in part, determinant(s) of the eliciting antigen, and some which did not^4,9. The MAB TB 68 has also been shown to bind to PPD¹⁰. Table 2 shows that all clones responded vigorously to the TB 68 antigen. The clones also responded to other antigenic preparations. Indeed Clone 13 also responded to a number of antigens which did not contain the eliciting antigen, particularly TB 72, which was purified in a similar fashion to TB 68. Clones 1, 2, 6 and 7 on the other hand proliferated only to those antigenic preparations, which contained TB 68 antigen i.e. Mycobacterium tuberculosis strain H37Rv. Mycobacterium bovis strain BCG and PPD. Not all clones responded identically. In particular. Clone 2 did not respond to PPD or to M bovis BCG antigenic preparations, indicating that this clone may recognise a different epitope of the antigen than either Clones 1, 6 or 7. In its native state the TB 68 antigen is large with a molecular weight of 120 K. On SDS treatment 4 major chains are found. Clone 2 may, therefore recognise a different chain to Clones 1, 6 and 7. All clones were responsive to IL2 (lymphocult T, lectin free, Biotest Folex Limited) alone, although to a lesser extent than antigen alone. Optimal proliferation was achieved in the presence of both antigen and IL2.

We have examined the intrinsic capacity of clone supernatant to increase H₂O₂ release ¹¹ in a human myeloid cell line, U937, as an indication of macrophage activating factor activity¹². U937 is a macrophage cell line that can be immunologically activated to express antimicrobial activity¹². Table 3 shows that all supernatants from clones with helper/ inducer phenotypes, irrespective of the pattern of antigen specificity in the clone proliferative response, increased H₂O₂ production. The clone supernatants were 10-fold more effective than a whole phytohaemaglutinin A-activated (PHA) spleen cell supernatant (SCF), indicating that the production of macrophage activating factor(s) (MAF) is an important constituitive function of these clones. This activity does not appear to be due to interferon!3. Clone 7 supernatant did not have MAF activity. This clone is currently being tested to determine whether it is functionally a suppressor or a cytotoxic T cell. If it is a suppressor T cell clone, this would suggest that this particular antigen may also stimulate the proliferation of T ceils involved in the regulation of the macrophage activating response. Preliminary experiments indicate that helper/inducer clones, cultured with U937 cells, were as good, or frequently better, than supernatants in increasing H₂O₂ release. Cloned cells as well as lymphokine could also enhance H₂O₂ release from conventional macrophages matured from peripheral blood monocytes , indicating that this activity was not dependent on using U937 cells. Sham supernantants , prepared from identical cultures with autologous feeders, but in the absence of cloned cells, had a small, but measurable effect on H₂O₂ production. This was not due to Interleukin 2, which even at 50% concentration did not increase H₂O₂ production above control levels .

The TB 68 antigen is not alone in stimulating MAF production by lymphocytes, a variety of stimuli have been used^14,15. However, using T cell cloning we have demonstrated that helper/inducer but not suppressor/cytotoxic T cells secrete MAF, although cells of both phenotypes specifically proliferate in vitro to TB 68 antigen. Previously lymphokine production and proliferation to soluble antigens were thought to be the function of distinct populations of T cells^16,17. Whilst the exact relationship between the release of H₂O₂ and microbicidal functions remains to be clarified, H₂O₂ appears to kill M microti in mouse macrophages¹⁸. Therefore, these T cell clones, either by cell to cell contact, or through lymphokines, may enhace the microbicidal function of the macrophage which is the main effector cell of immunity to Mycobacterium tuberculosis.

We have shown that a monoclonal antibody having specificity for the T3 cell surface antigen can be used to promote maintenance of a cloned human T cell culture as an alternative to an appropriate antigen and feeder cells . LEGENDS

Table 1

Slides of T cell clones were prepared on a cytospin centrifuge (Shandon). Glass slides were incubated with MABs as specified. Binding of MAB to the cells was detected by immunoenzymatic reaction¹⁹, using alkaline phosphatase conjugated to rabbit and mouse IgG (Sigma, London) with substrate; Naphthol-AS-MX Phosphate (Sigma!, London) dissolved in N,N,- DimethyIformamide (Sigma, London) in 0.1 M Tris buffer, pH 7.2 with fastred. Slides were examined by light microscopy and cells were counted and the number staining were expressed as a percentage of the total cells viewed.

Table 2

10⁵ T cells from 5 T cell clones were stimulated with TB 68 antigen (0.0117 g/ml), or other antigens at specified and previously determined optimal concentrations. Other antigens were either a similarly purified protein antigen binding to MAB 72 or mycobacterial pressates⁴ (the non-particulate fraction of a bacterial homogenate passed through a press)⁴. Cloned cells were cultured with antigen and 10⁴ autologous, irradiated (3.000 Rads 35 Cs source) PBL for 48 hours in microterasaki plates in humidified chambers in 5% CO₂/air. Stimulation was determined by incorporation of ³H Thymidine as described previously²⁰. The results are expressed as the meas of triplicate counts. Table 3

U937 cells at 10⁶ cells/tube were cultures in RPMI 1640 medium (Gibco Biocult Limited). 10% foetal calf serum in the presence and absence of various concentrations of cloned T cell supernatants, sham control (a supernatant prepared from identical cultures but in the absence of cloned T cells) and spleen cell factor (SCF) produced by mitogen activated human spleen cells cultured with PHA for 72 hours. Medium and clone supernatants and SCF in U937 cultures were replaced at 24 hour intervals. At 72 hours the cells were rinsed in Hanks balanced salt solution. Peroxide release was then determined in the presence of phorbol myristate acetate (1 mg/ml) by the method of Jackett et al. , 1981¹¹. Results are given as nmol of H₂O₂ per 10⁶ U937 cells after subtraction of H₂O₂ production in the absence of supernatants or SCF.

References 1) Mackaness. GB. J Exp Med 129, 973-992 (1969) 2) Hahn, H and Kaufmann, S H E. Rev Infect Dis 3, 1221-1250 (1981) 3) Rees, A D M et al. Manuscript in preparation. 4) Coates, A R M; Hewitt, J; Allen, B W; Ivanyi, J and Mitchison, D A. Lancet i i , 167-169 (1981) 5) Lowrie. D B in the Biology of Mycobacteria Volume 2, 235-278. Academic Press (1983) 6) Matthews, R C et al . Manuscript in preparation. 7) Lamb, J R; Eckels, D D; Lake, P; Johnson, A H; Hartzman, R J and Woody. J N. J Immunology 128. 233-238 (1982) 8) Evans, R L; Wall, D W; Phatsoucas, C D; Siegal, F P; Fikrig, S M; Jesta, C M and Good, R A. Proc Nat 1. Acad Sci USA 78. 544-548 (1981) 9) Hewitt, J; Coates, ARM; Mitchison, D A and Ivanyi, J. J Immunol Methods 55, 205-211 (1982) 10) Harboe. Personal communication 11) Jackett, P S; Andrew, P W; Aber, V R and Lowrie, D B. J Exp Path 62, 419-428 (1981) 12) Wing, E J; Koren. H S; Fischer, D G; Kelley, V. J Reticuloendoth Soc 29, 328-330 (1981) A 13) Andrew, P W et al. Manuscript in preparation. 14) Lohmann-Matthes, M L; Ziegler, F G and Fischer, H. Eur J Immunol 3, 56-58 (1975) 15) Pfizenmaier, K; Trostmann, H; Rollinghoff, M and Wagner, H. Immunology 29, 967-976 (1975) 16) Woody, J N; Ahmed, A; Knudsen, R C; Strong, D M and Sell, K W. J Clin Invest 55, 956-966 (1975) 17) Evans, R L; Breard, J M; Lazarus, H; Schlossman, S F and Chess, L. J Exp Med 145 (1977) 18) Walker, L and Lowrie, D B. Nature 293. 69-70 (1981) 19) Moir, D J; Ghosh, A K; Abdulaziz, Z; Knight, P M and Mason, D Y. Brit J Haem 55, 395-410 (1983) 20) O'Brien, J; Knight, S; Quick, N A; Moore, E H and Platt, A S. J Immunol Methods 27, 119-223 (1979)

Claims

1. A macrophage activating factor characterised in that the macrophage activating factor is antigenically distinct from interferon-γ.

2. A macrophage activating factor according to claim 1 which is antigenically the same as or similar to a macrophage activating factor produced by cloned human T cell line TB 68.2.1, TB 68.2.2 or TB 68.2.6.

3. A cloned human T cell line designated TB 68.2.1 deposited under accession number 85030401, mutants and derivatives thereof.

4. A cloned human T cell line designated TB 68.2.2 deposited under accession number 85030402, mutants and derivatives thereof.

5. A cloned human T cell line designated TB 68.2.6 deposited under accession number 85030403, mutants and derivatives thereof.

6. A process for producing macrophage activating factor according to claim 1 comprising the steps of culturing a cloned human T cell line which has been stimulated prior to cloning with a TB 68 antigen and isolating the macrophage activating factor from the culture supernatant.

7. A process according to claim 6 wherein the cloned human T cell line is cultured in a culture medium which includes a cell growth factor.

8. A process according to claim 7 wherein the cloned human T cell line is cultured in the presence of an antigen which stimulates secretion of macrophage activating factor and autologous feeder cells.

9. A process according to claim 7 wherein the cloned human T cell line is cultured in the presence of a monoclonal antibody having specificity for the T3 cell surface antigen.

10. A pharmaceutical composition comprising macrophage activating factor according to claim 1 or 2 and a pharmaceutically acceptable excipient.